Ring Oscillator RNG Calibration for Stable Bit Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ring oscillators used in random number generators experience significant variations in random bit generation rate due to changes in temperature, power, and voltage, leading to increased dispersion and security vulnerabilities.
Innovation Solution
A random number generator system that includes a ring oscillator, an inversion selecting circuit, and a controller, allowing for frequency monitoring and calibration by switching between modes to optimize performance and bit rate, using a phase inverter in one mode for measurement and disabling it in another for generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a ring oscillator is used to generate random numbers, then random number generation capability is provided, but the random bit generation rate varies significantly with temperature, power, and voltage changes
Solution Approach 1:
The patent implements a feedback mechanism by measuring the frequency of the ring oscillator and using this measurement to control the phase inverter operation. The controller monitors the oscillator frequency and adjusts the phase inverter accordingly to maintain stable random bit generation rate despite PVT variations.
Solution Approach 2:
The patent changes the operational parameters of the ring oscillator by selectively enabling or disabling the phase inverter based on measured frequency conditions. This parameter change allows the system to adapt to temperature, power, and voltage variations while maintaining stable random number generation performance.
2Reliability
If frequency monitoring and calibration is implemented, then random performance and bit rate are optimized, but device complexity increases
Solution Approach 1:
The patent makes the phase inverter serve multiple functions: it acts as a normal signal inverting element during random number generation and as a frequency monitoring mechanism when enabled. This multi-functionality reduces the need for separate monitoring circuits, thereby limiting the increase in device complexity.
Solution Approach 2:
The ring oscillator system performs self-diagnosis and self-calibration by using its own output signal to monitor its frequency. The system automatically adjusts its operation based on internal measurements without requiring external calibration equipment, thereby minimizing additional complexity.
3Measurement precision
If a phase inverter is used for frequency measurement, then oscillator frequency can be monitored, but resources and operational overhead increase
Solution Approach 1:
The patent implements periodic frequency measurement rather than continuous monitoring. The controller periodically enables the phase inverter to measure frequency and then disables it for random number generation. This periodic action reduces operational overhead and resource consumption compared to continuous monitoring.
Solution Approach 2:
The patent dynamically switches the phase inverter between two operational states: enabled for frequency measurement and disabled for random number generation. This dynamic operation allows the system to perform measurements only when necessary, reducing overall operational overhead and resource usage.
Data Source
AI summary
A random number generator includes a ring oscillator, an inversion selecting circuit, and controller. The ring oscillator includes an inverter chain having at least one inverter and generates an output signal. The inversion selecting circuit controlling a phase inverter configured to invert a signal of the inverter chain. The controller is configured to operate the inversion selecting circuit to provide an output of the first phase inverter to the inverter chain during a first operation mode to measure a frequency of the ring oscillator and operate the inversion selecting circuit to not provide the output of the phase inverter during a second operation mode for generating a random number.


